@inproceedings{towardslongdelayedsparsitylearninga, title = {Towards Long-delayed Sparsity: Learning a Better Transformer through Reward Redistribution}, author = {Tianchen Zhu and Yue Qiu and Haoyi Zhou and JianXin Li}, year = {2023}, booktitle = {journal 2023 2}, url = {https://www.ijcai.org/proceedings/2023/0522.pdf}, }